This proposal examines the physiological function of a novel class of cAMP-regulated proteins, the exchange proteins activated by cAMP (Epacs) and tests the novel hypothesis that each isoform (Epac1 and ERpac2) carries out distinct physiological functions in the cell. This family of proteins binds cAMP directly and represents the major intracellular target of cAMP other than PKA itself. Our studies of Epac activation of small G protein Rap1 have demonstrated that subcellular localization of exchangers can dictate the choice of effectors utilized by the G proteins they activate (Wang et al. 2006; Liu et al. 2008). We have identified distinct targeting domains in both Epac 1 and Epac2 called Ras association (RA) domains (Liu et al. 2008). We propose that these domains allow specific small G proteins to link to Epac1 and Epac2, respectively, to the activation of distinct subcellular pools of Rap1. We propose that the RA domains of Epac1 and Epac2 link cAMP signaling to the small G proteins, Ran and Ras, respectively. This can explain how Ran couples to Rap1 (via the Epac1 exchanger) and how Ras can couple to Rap1 (via the Epac2 exchanger). This model and its physiological consequences for both Epac1 and Epac2 are examined in this proposal. We have discovered that Epac1 binds to the small G protein Ran at the nuclear pore. We will test the hypothesis that the RA domain of Epac1 couples Ran-GTP to Rap1 to regulate the transport of proteins in and out of the nucleus. This represents the first link of Ran to intracellular signaling cascades and identifies a novel function for both Epac1 and Rap. The significance of this finding is that the nuclear translocation of proteins via the Ran cycle governs cell growth and differentiation, and its dysregulation is a hallmark of many cancers. We show that Epac1 regulates the nuclear functions of the oncoprotein 2-catenin. We propose that this and related functions of Epac1 may explain some of the anti-proliferative effects of cAMP. In contrast to Epac1, we propose that Epac2 binds to activated Ras at the plasma membrane to link cAMP and Ras to Rap1 activation. This model of Epac2 function as a coincidence detector for signals through Ras and cAMP has significant implications for diabetes, as Epacs are now known to be targets of the sulfonylureas. We propose here for the first time a model that Ras and cAMP signals converge on Epac2 to potentiate insulin release. If true, this would establish an innovative approach to diabetes involving combining Ras activators with both insulinomimetic hormones and the sulfonylurea drugs.